Update modeling_gpt2vision.py

modeling_gpt2vision.py

@@ -92,6 +92,10 @@ class GPT2Vision(PreTrainedModel):
         batch = {"image": [image], "text": prompt}
         encoded_batch = self.tokenize_encode(batch, self.device)
         inputs_embeds, attention_mask, input_ids = self.preprocess_inputs(encoded_batch)
+
+        print("input_embeds",inputs_embeds.size())
+        print("attention_mask",attention_mask.size())
+
         output_sequences = self.language_model.generate(
             inputs_embeds=inputs_embeds,
             attention_mask=attention_mask,